Search Results (287)

In satellite-denied environments, conventional localization methods struggle with rapid pose initialization due to the absence of global positioning data. To address this challenge, this study presents a high-precision pose initialization framework based on a Siamese Neural Network (SNN) and multi-stage probabilistic registration localization. First, the SNN improved by Convolutional Block Attention Module (CBAM) matches features between real-time radar point clouds and prior map slices, producing candidate positions based on similarity scores. Then, Adaptive Monte Carlo Localization (AMCL) performs probabilistic matching among these candidates to identify the correct slice and refine the position accuracy from tens of meters to meter-level, along with an approximate orientation estimate. Finally, the Normal Distributions Transform (NDT) is applied for point cloud registration, achieving centimeter-level pose estimation. The proposed method is evaluated on self-collected medium-scale and large-scale maps. Experimental results show that the SNN effectively identifies the correct map slice, which is further refined by AMCL and NDT to achieve centimeter-level position accuracy and sub-degree orientation accuracy. The multi-stage method achieves localization success rates of 99% on both 200 × 100 m and 300 × 200 m regions, with distance RMSEs of 0.175 m and 0.348 m, and orientation RMSEs of 0.149° and 0.437°, respectively. Evaluations on the KITTI dataset further demonstrate robust initialization performance in complex outdoor environments. The proposed framework provides a reference for high-precision pose initialization in large-scale satellite-denied scenarios. Full article

Accurate identification of Chinese sensitive words is critical for maintaining online information security. However, this task faces three technical challenges: (1) high contextual dependency causing semantic ambiguity; (2) adversarial variations (e.g., homophones, character splitting) that evade exact matching; and (3) scarcity of high-quality annotated samples in complex scenarios, leading to few-shot distribution characteristics. To address these challenges, we propose a Branch Transformer Mutual-Information Calibration (BTMC) network. Specifically: (i) to capture multi-level, cross-dimensional semantic interactions despite limited data, we design a branch-based Transformer structure that aligns and fuses features across different semantic dimensions; (ii) to establish context channels between global and local semantics under few-shot conditions, we introduce a global-local interactive fusion mechanism that enhances focus on core semantics; (iii) to improve discriminability of complex semantic patterns, we propose a semantic calibration regularization mechanism that reweights features and balances information distribution. Experimental results on a newly constructed Chinese sensitive words dataset (45,623 sentences, four categories) demonstrate that BTMC achieves average F1-scores of 0.9715 (Politics and Violence), 0.9683 (Rudeness and Vulgarity), 0.9704 (Drugs and Gambling), and 0.9531 (Others), outperforming state-of-the-art baselines by 10–15% relative improvement. The code and dataset will be made publicly available. Full article

Accurate and efficient 3D reconstruction of trees is of paramount importance for studying forest spatial structures and dynamic resource patterns, optimizing forest management, protecting environments, and analyzing carbon cycles. Currently, Light Detection and Ranging (LiDAR) remains the dominant method for generating 3D models of forest scenes. However, with advancements in computer vision, photogrammetry has emerged as a crucial tool for forest inventory and 3D reconstruction due to its cost-effectiveness. Nevertheless, in practical forestry applications, traditional photogrammetry often suffers from low reconstruction efficiency and poor quality during feature extraction and matching. These issues stem from the complex structure of forest scenes, severe occlusion, and repetitive texture patterns. To address these challenges, this paper proposes an improved 3D tree reconstruction approach based on images, integrating deep learning-based methods. In the sparse reconstruction stage, we utilize the ALIKED (A LIghter Keypoint and descriptor Extraction network with Deformable transformation) algorithm and construct an image pyramid to extract multi-scale robust features. Furthermore, by combining the LightGlue matching algorithm with a neighborhood search constraint strategy, we enhance the stability of camera pose recovery while reducing redundant computations. Experimental results demonstrate that our method outperforms traditional algorithms in both accuracy and robustness regarding image matching. Compared to baseline models, the proposed approach increases the number of feature points by approximately 50% with a more widespread distribution, improves matching accuracy by 4% to 8%, and achieves a 100% image registration rate. Consequently, under the condition of maintaining equivalent re-projection errors, the subsequent sparse point clouds exhibit an average track length increase of 0.6 to 1.4 and a density increase of up to 1.2 times. Notably, this method effectively mitigates artifacts and spurious reconstructions caused by pose drift in forest photogrammetry. Full article

The sixth generation of mobile networks (6G) represents a paradigmatic shift in the conception of wireless communication systems, where Artificial Intelligence (AI) is not integrated as an additional feature but is conceived as a native and fundamental component of the physical layer (PHY). This paper presents a comprehensive survey of the state of the art in AI-native physical layer for 6G, synthesizing approximately 100 references from the period 1948–2025. The survey systematically covers 5 main PHY components (channel coding, channel estimation, signal detection, beamforming, and semantic communications) and analyzes 8 AI architectural families (autoencoders, CNN, RNN/LSTM, Transformers, GNN, GAN, Diffusion Models, and Foundation Models), addressing theoretical foundations, proposed architectures, learning algorithms, implementation challenges, and future research directions. A rigorous mathematical framework underpinning these developments is presented, including optimization formulations, convergence analysis, and theoretical performance characterization. Published results from the literature demonstrate that AI-native physical layer can improve conventional performance metrics and enable emerging capabilities essential to 6G, such as semantic communications, predictive environmental adaptation, and operation in previously inaccessible computational complexity regimes. However, such gains are conditional on adequate training resources, robust channel-matched data, and careful consideration of known limitations including generalization across channel distributions, sample inefficiency, model interpretability, and hardware implementation constraints—all of which are critically analyzed in this survey. A reproducible proof-of-concept benchmark further confirms that, under severe resource constraints, autoencoder-based codes currently underperform conventional schemes, highlighting the gap between theoretical potential and practical deployment readiness. Full article

As urbanization picks up pace and the public demand for security keeps climbing, video surveillance systems have emerged as a vital tool for maintaining social stability and safeguarding public safety. Person Re-Identification (Re-ID), as one of the core technologies in intelligent monitoring, mainly aims to accurately match pedestrian identities across cameras without overlapping fields of view. However, in practical applications, occlusion remains a primary challenge that severely degrades Re-ID performance. Especially in high-density crowds, pedestrians are often partially or completely obscured by other objects or individuals, resulting in incomplete image information and impaired feature representation, which significantly reduces recognition accuracy and reliability. Aiming at the problems of excessive reliance on external pose estimation models and asymmetric information matching in occluded Re-ID, this paper proposes a transformer-based pedestrian background decoupling network. The algorithm achieves foreground–background separation and multi-scale feature matching through the synergy of three modules. Meanwhile, a two-stage training strategy is adopted: the first stage optimizes the decoupling module to ensure clean feature separation, while the second stage jointly fine-tunes the correlation module to enhance matching accuracy. Extensive experimental results show that the proposed algorithm outperforms existing methods. Full article

Xinyu pear is one of the important pear cultivars in China. Owing to its rich nutritional composition, high quality, and distinctive flavor, it is highly favored in the market. In this study, near-infrared spectroscopy was employed to determine the soluble solid content (SSC) of Xinyu pears. To investigate the influence of spectral preprocessing on SSC prediction, near-infrared spectra of two batches of Xinyu pear samples were collected using the same portable spectrometer under different acquisition parameters, resulting in differences in spectral bands. A linear interpolation method was introduced to the first batch to generate a new dataset to match the dimensionality of the second batch, and a total of three datasets were used. Five preprocessing methods, including moving average smoothing (MA), standard normal variate transformation (SNV), multiplicative scatter correction (MSC), first derivative (D1), and second derivative (D2), together with three regression models, namely partial least squares regression (PLSR), support vector regression (SVR), and convolutional neural network (CNN), were systematically evaluated and compared in terms of predictive accuracy. Overall, PLSR achieved the best prediction performance, followed by CNN and SVR. Certain differences in model performance were observed among the three datasets. In general, MA exhibited the best overall performance across different datasets and models. Although SNV and MSC were slightly inferior to MA, they showed relatively stable predictive accuracy. By contrast, prediction models based on derivative spectra generally performed poorly. To further exploit the complementary information contained in differently preprocessed spectra, a four-branch CNN model was constructed using raw spectra, MA-preprocessed spectra, SNV-preprocessed spectra, and MSC-preprocessed spectra as separate inputs. Based on the fused features extracted by the CNN, PLSR and SVR models were subsequently developed. The prediction correlation coefficients of the feature-fusion CNN model on the prediction sets of the three datasets were 0.8811, 0.8259, and 0.7064, respectively. For the original datasets of the first and second batches, the feature-fusion model outperformed all single-preprocessing models. For the dataset generated by linear interpolation, the predictive performance of the feature-fusion strategy was comparable across the three models; specifically, its accuracy in SVR exceeded that of all single-preprocessing models, while its accuracies in CNN and PLSR surpassed those of most preprocessing methods. These results demonstrate that integrating feature information from spectra subjected to different preprocessing methods is a feasible strategy for improving prediction accuracy. This study provides an effective reference for SSC prediction in Xinyu pears based on portable spectrometers. Full article

Flat label supervision often constrains multi-label image classification, as it struggles to fully capture inherent label dependencies. It provides limited guidance to the hierarchical features that naturally emerge in Vision Transformers. To address this structural misalignment, we propose Dynamic Topic-based Hierarchical Prompt Learning (DyT-HPL). Instead of relying on predefined and fixed label graphs, DyT-HPL utilizes offline hierarchical clustering to construct multi-granularity semantic priors, from which hierarchical prompts are dynamically retrieved. A frozen visual query branch generates stable semantic queries, which are then used to retrieve discrete prompts from constructed coarse-mid-fine prompt pools. These hierarchical prompts are adaptively injected into different network depths, ensuring that semantic guidance with different abstraction levels is introduced at the most suitable architectural stages. To maintain stable routing and prevent prompt mode collapse, we jointly optimize the architecture with asymmetric classification, surrogate matching, and intra-pool diversity losses. This tripartite design promotes a diverse prompt space and isolates routing updates from final predictions. Comprehensive experiments on MS-COCO, NUS-WIDE, and Corel5k demonstrate that DyT-HPL achieves consistent and favorable performance across diverse settings, highlighting the value of hierarchical semantic guidance with different abstraction levels. Full article

Fourier representations play a central role in operator learning for partial differential equations and are increasingly being explored in quantum machine learning architectures. The classical fast Fourier transform (FFT), particularly in its Cooley–Tukey decomposition, exhibits a structure that naturally matches continuous-variable quantum circuits. This correspondence establishes a direct structural isomorphism between the Cooley–Tukey butterfly network and Gaussian photonic gates, enabling the FFT to be realized as a native optical computation in continuous-variable quantum computing. Building on this observation, we introduce a continuous-variable Quantum Fourier Layer (CV–QFL) based on a bipartite Gaussian encoding and a Cooley–Tukey quantum Fourier transform, enabling exact two-dimensional spectral processing within a Gaussian photonic circuit. We test the CV–QFL on two representative tasks: spectral low-pass filtering and Fourier-domain integration of the heat equation. In both cases, the results match the classical reference to machine precision. More broadly, this work lays the foundation for continuous-variable approaches to quantum scientific computing and for the development of native spectral architectures in quantum machine learning. Full article

Pore-throat structures in a carbonate reservoir were classified into ten petrophysical facies representing coarse, medium, or fine throat types based on Mercury Injection Capillary Pressure (MICP) data from 77 core samples, directly reflecting distinct flow capacities. Using Nuclear Magnetic Resonance (NMR) data from 20 samples, an artificial neural network (ANN) model was developed with four conventional logs, namely Gamma Ray (GR), Deep Laterolog Resistivity (RD), Density (DEN), and Compensated Neutron Log (CNL), as inputs to predict the T₂ spectrum continuously. A cumulative pore-throat size distribution matching method was then used to transform predicted T₂ spectra into capillary pressure curves. The resulting pore-throat parameters show excellent agreement with core measurements, with relative errors for key parameters—such as median pore-throat radius (R₅₀) and sorting coefficient (Sp)—below 15%. This approach extends discrete core data to continuous wellbore profiles, enabling pore-throat prediction and facies classification in intervals lacking MICP data. It effectively identifies dominant flow channels and tight interlayers, with facies validated by thin-section petrography, providing a robust basis for evaluating highly heterogeneous carbonate reservoirs. Full article

Occluded person re-identification (Re-ID) aims to accurately match occluded pedestrian images against complete gallery images captured across multiple cameras, a task that is critical to public security and intelligent surveillance systems. Existing graph neural network (GNN)-based methods typically assign uniform aggregation weights to all nodes, failing to reflect the inherent reliability difference between visible and occluded body regions, which allows noise from low-confidence nodes to propagate freely and corrupt the final pedestrian representation. To address this, we propose the Self-Information-Driven Gated Graph Convolutional Network (SI-GCN). Keypoint detection confidence scores are transformed into logarithmic self-information measures as uncertainty priors for a learnable gating mechanism. The proposed SIG module enables visible nodes to dominate information diffusion while occluded nodes absorb more from neighbors, achieving efficient feature updating. A dynamic confidence calibration (DCC) strategy further synchronizes node reliability estimates with feature evolution across successive GCN layers. Extensive experiments on six public benchmarks covering occluded, partial, and holistic Re-ID scenarios demonstrate that SI-GCN achieves state-of-the-art performance, with Rank-1 accuracy and mAP improvements of 1.2% and 0.9%, respectively, over the strongest baseline on the Occluded-REID dataset, demonstrating its strong potential for deployment in real-world public security and urban surveillance applications where occlusion is pervasive. Full article

Reliable navigation is critical for UAVs operating in GNSS-denied environments, where conventional Inertial Navigation System/Global Navigation Satellite System (INS/GNSS) integrated navigation struggles to meet the requirements of high-reliability and long-endurance missions. As a passive and autonomous approach, terrain-aided navigation (TAN) offers strong concealment and a high degree of autonomy. However, most existing TAN methods rely on handcrafted features, which limit their ability to fully exploit multi-level terrain information, while sensitivity to elevation noise and attitude variations further degrades matching accuracy and robustness. To address these issues, this paper proposes a GNSS-denied UAV terrain matching navigation method based on an autoencoder network with contrastive learning. A Global–Local Dual-branch Feature Extraction Network (GL-DualNet) is designed to combine the local detail extraction capability of CNNs with the global dependency modeling ability of the Swin Transformer, enabling effective multi-scale terrain representation. In addition, an Autoencoder Contrastive Learning Model (ACLM) is developed to jointly optimize reconstruction and contrastive objectives, enabling unsupervised learning of terrain features with improved discriminability and robustness against noise and rotational disturbances. Experiments on a public terrain dataset show that the proposed method outperforms conventional terrain matching approaches under different noise levels, rotational disturbances, and search ranges, demonstrating its effectiveness and robustness for UAV navigation in complex GNSS-denied environments. Full article

This paper presents a theoretically grounded integration of deterministic Q-learning with relational game theory (QLRG) for efficiently identifying minimal winning coalitions in Online Social Networks (OSNs). We address the fundamental challenge that coalition formation is NP-hard under traditional approaches by leveraging structural properties of relational dependencies and Armstrong’s axioms to transform the problem into one solvable in polynomial time. Our framework reduces the state space from exponential O(2ⁿ) to O(n²) through a sufficient statistic representation based on coalition size, follower reach, and terminal status, while achieving O(n⁴) time complexity under deterministic, static, and sufficiently symmetric influence structures. The QLRG framework introduces three critical innovations: (1) a principled agent selection mechanism derived directly from the Q-function that eliminates heuristic weight tuning; (2) a formal Boost action defined through temporal closure operators that captures influence spread dynamics; and (3) a constrained MDP formulation that enforces relational consistency through action elimination rather than penalty terms. We prove that the Bellman optimality operator forms a contraction mapping, guaranteeing deterministic convergence to optimal policies with established rates of O(1/√k) for decreasing learning rates or linear convergence up to bias for constant rates. To bridge the gap between this idealized model and the asymmetry inherent in real OSNs, we further develop a cluster-based sufficient statistics approach. By partitioning the network into communities with bounded internal variation, we relax the global symmetry requirement while preserving polynomial state space complexity, and obtaining a single within-community swap changes the optimal Q-value by at most ${\displaystyle \frac{{\epsilon }_i}{1-\gamma }}$, which is a local Lipschitz continuity result. The implications of this are both theoretical and practical, and they form the bedrock for relaxing the global symmetry assumption in the QLRG framework. Empirical validation on synthetic networks satisfying the symmetry assumption demonstrates that QLRG consistently identifies minimal winning coalitions matching the optimal solutions found by exhaustive search, while operating with polynomial-time complexity. Unlike conventional approaches, our framework simultaneously satisfies four critical properties: deterministic convergence, policy optimality, minimal coalition identification, and computational tractability. The work bridges computational social science and operations research, providing a mathematically rigorous foundation for strategic decision-making in influencer marketing and coalition formation. While the framework requires symmetry assumptions that may only hold approximately in real-world OSNs, it establishes an idealized baseline for future extensions addressing stochasticity, dynamics, and partial observability. This research represents a paradigm shift from empirical improvements to theoretically grounded convergence guarantees for coalition formation problems, demonstrating how structural mathematical insights can transform intractable problems into efficiently solvable ones without sacrificing solution quality. Full article

Refined decision-making of the design intent is a key factor affecting the iterative design of complex equipment products. While current research on design intent decision-making generally emphasizes methodological innovation, it often neglects the individualized and fuzzy expressive characteristics of cognitive agents, as well as the actual status of the research object. This oversight leads to uncertainty in both design intent and design outcomes. To address these issues, in this paper, a refined decision-making method for the multimodal fuzzy design intent of complex products based on noncooperative–cooperative game serialization is proposed. First, through scenario analysis, the fuzzy design intent evaluation process of different cognitive agents is transformed into a cooperative game model based on a fuzzy network, achieving a preliminary assessment of design intent. On this basis, a noncooperative game-based refined matching and decision-making model for design intent across different dimensions is constructed, thereby completing the final design intent decision-making for a specific product model. Finally, the proposed method is applied to the design intent decision-making process of a CKA6180 CNC machine tool, yielding the conclusion that the two design intents of “good protective performance” and “grand appearance” should be prioritized, thereby verifying the practicality and effectiveness of the method. The analysis of the results reveals the following: ① The application of scenario analysis theory enables a more comprehensive and precise characterization of the design intents of different cognitive agents; ② The construction of a model combining a fuzzy network with a cooperative game facilitates a more complete representation and evaluation of multimodal fuzzy design intent data; ③ The integration of a refined design concept with a noncooperative game model leads to more definitive design intent decision outcomes, thereby reducing the “disturbance” of experience dependence in the early design phase and consequently enhancing subsequent design satisfaction. Full article

Background: Prostate cancer (PCa) is one of the most prevalent malignancies in men, and active surveillance (AS) is the recommended management strategy for low- and favourable intermediate-risk disease. Predicting which patients will progress during AS remains a clinical challenge. MRI-derived radiomics has shown promise for risk stratification, but conventional machine learning approaches treat radiomic features as independent variables and may not capture the complex inter-feature dependencies within imaging data. This study evaluates the application of Synolitic Graph Neural Networks (SGNNs) to baseline MRI-derived radiomic features for predicting prostate cancer progression on active surveillance. Methods: We studied 343 AS patients (73 progressors, 270 non-progressors) from a single-centre cohort prospectively enrolled between 2013 and 2019 and retrospectively analysed. Seventy-two radiomic features were extracted from baseline 3T MRI (T2-weighted imaging and apparent diffusion coefficient maps), together with three clinical variables (prostate volume, PSA, PSA density). The SGNN pipeline transformed each patient’s feature profile into a weighted graph encoding pairwise feature relationships via logistic regression classifiers trained within each cross-validation fold. GCN and GATv2 architectures were evaluated with multiple sparsification strategies and compared against Gradient Boosting, SVM, Random Forest, and logistic regression using 5-fold stratified cross-validation. Results: Among conventional methods, Gradient Boosting achieved the highest ROC-AUC (0.634 ± 0.080). The SGNN pipeline with GATv2, confidence-based sparsification (p = 0.8), and extended node features incorporating graph centrality measures achieved the best performance (ROC-AUC = 0.699 ± 0.044), an absolute improvement of 0.065 over the best conventional method. The addition of topological node features consistently improved performance by 3–5% across configurations. GATv2 outperformed GCN in matched settings. Conclusions: As a proof of concept, the SGNN framework achieved the highest mean ROC-AUC among the evaluated single-timepoint approaches, though results require confirmation in independent external cohorts. By encoding inter-feature relationships as patient-specific graphs, SGNN offers a complementary modelling paradigm for radiomic data in clinical oncology. Future work should incorporate longitudinal data and graph explainability methods. Full article

To address the issues of difficult heterogeneous image registration and low segmentation accuracy caused by the severe lack of illumination and significant modal differences in concrete cracks in extremely dark environments, this paper proposes a two-stage processing framework of registration–fusion first, and decoupling–segmentation later. In the registration and fusion stage, a registration algorithm based on morphological priors and multi-level quadtree spatial constraints is designed. This approach transforms the problem from pixel grayscale matching to spatial topological matching, achieving a feature fusion of high infrared saliency and high visible light sharpness. In the segmentation stage, a Latent Frequency-Decoupled Topological Network (LFDT-Net) is proposed. It utilizes Discrete Wavelet Transform (DWT) to achieve high-fidelity frequency decoupling of the low-frequency infrared backbone and the high-frequency visible light edges. Furthermore, a Cross-Frequency Guidance Module is utilized to eliminate double-edged artifacts, and a skeleton-aware topological loss function is introduced to constrain the topological integrity of the cracks. Experimental results on a self-built heterogeneous multi-modal crack dataset demonstrate that the proposed method significantly outperforms existing mainstream methods in registration accuracy, fusion quality, and segmentation accuracy. Achieving a mean Intersection over Union (mIoU) of 81.7%, the method effectively suppresses background noise in dark environments and precisely restores the microscopic edges and continuous topological structures of faint cracks. Full article